This grant application is for a second renewal of our group?s key NIH grant that supports development of optogenetic tools -- microbial opsins that enable safe, temporally precise, and high-magnitude control of neural activity in neurons in awake behaving mammals and other species of importance in neuroscience. Since our grant was first awarded in 2010, it has supported the development of optogenetic tools such as Arch (the first optogenetic neural silencer to result in ~100% optogenetic silencing of neural activity in awake behaving mice), ArchT (a 3x more light-sensitive relative of Arch), Chronos (an ultrafast optogenetic activator, used in contexts where speed is essential), Chrimson (the most redshifted optogenetic activator, useful for activation of large volumes of brain tissue as well as avoiding behavioral artifacts in Drosophila), Jaws (the most redshifted optogenetic silencer), SoCoChR (which enables single-cell, single-spike resolution optogenetics) and ChromeQ (a potassium- and sodium-selective optogenetic activator), resulting in 50 peer reviewed papers, and resulting in wide distribution of next-generation optogenetic tools throughout neuroscience. To date, we have primarily used genomic search to discover novel opsins, mining public and private databases to identify new candidates. Having screened through a large number of genomic resources to identify these molecules, however, one concern is that there are diminishing returns, and that some goals will not be met purely through genomic search, or even structure-guided site-directed mutagenesis. Directed evolution, which sifts through a large number of mutants of a parent gene to identify versions improved towards some goal, offers hope, but has not been applied to optogenetic tools due to the difficulty of performing directed evolution in mammalian cells (essential, since optogenetic tools that express well in cells commonly used in directed evolution, such as E. coli, do not express well in mammalian cells, and evolving optogenetic tools in such cells would likely de- optimize them for expression in mammalian cells), and the difficulty of performing multidimensional directed evolution (essential, because we need to optimize optogenetic tools towards multiple goals ? for example, localization, spectrum, and magnitude ? and optimizing too much along one axis will de-optimize the tool along other axes). We here propose to develop a directed evolution approach for optogenetic tool engineering (Aim 1), and apply it to several longstanding open needs in optogenetics: the creation of redshifted and blue spectrum-trimmed optogenetic activators, Aim 2; the creation of multiphoton-optimized silencers, Aim 3; and the optimization (by developing and applying automated patch clamp technology) of kinetics and ion selectivity, aiming to improve optogenetic tool kinetics for the aforementioned optogenetic tools as well as potassium conductances of light-gated potassium channels (Aim 4). We aim to deliver to the neuroscience community a powerful toolbox of optogenetic controllers of widespread utility, and to disseminate them freely throughout the research world.